365天深度学习训练营-第P9周：YOLOv5-Backbone模块实现

• 本文为365天深度学习训练营 中的学习记录博客
• 参考文章：Pytorch实战 | 第P9周：YOLOv5-Backbone模块实现(训练营内部成员可读)
• 原作者：K同学啊|接辅导、项目定制

一、课题背景和开发环境

第P9周：YOLOv5-Backbone模块实现

• 难度：夯实基础⭐⭐
• 语言：Python3、Pytorch

要求：

• 本次我将利用YOLOv5算法中的Backbone模块搭建网络，后续理论部分介绍将在语雀以及公众号（K同学啊）中详细展开，本次内容除了网络结构部分外，其余部分均与上周相同。
• YOLOv5是目标检测算法，是否可以尝试将其网络结构用在目标识别上，或进行改进形成一个全新的算法（类似之前介绍过的VGG1-6）。如果效果不错的话，还可以搞一篇期刊文章出来~

分享一张 K同学啊 绘制的YOLOv5_6.0版本的算法框架图，希望它可以有助于你完成本次探索~

---

开发环境

• 电脑系统：Windows 10
• 语言环境：Python 3.8.2
• 编译器：无（直接在cmd.exe内运行）
• 深度学习环境：Pytorch
• 显卡及显存：NVIDIA GeForce GTX 1660 Ti 12G
• CUDA版本：Release 10.2, V10.2.89（cmd输入nvcc -V或nvcc --version指令可查看）
• 数据：K同学啊的百度网盘

---

二、前期准备

1.设置GPU

如果设备上支持GPU就使用GPU,否则使用CPU

import torch
import torchvision

if __name__=='__main__':
    ''' 设置GPU '''
    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
    print("Using {} device\n".format(device))

Using cuda device

---

2.导入数据并划分数据集

import os
import PIL
import random
import pathlib
import warnings
import numpy as np
import matplotlib.pyplot as plt

''' 读取本地数据集并划分训练集与测试集 '''
def localDataset(data_dir):
    data_dir = pathlib.Path(data_dir)
    
    # 读取本地数据集
    data_paths = list(data_dir.glob('*'))
    classeNames = [str(path).split("\\")[-1] for path in data_paths]
    
    # 关于transforms.Compose的更多介绍可以参考：https://blog.csdn.net/qq_38251616/article/details/124878863
    train_transforms = torchvision.transforms.Compose([
        torchvision.transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
        # torchvision.transforms.RandomHorizontalFlip(), # 随机水平翻转
        torchvision.transforms.ToTensor(),          # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
        torchvision.transforms.Normalize(           # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
            mean=[0.485, 0.456, 0.406], 
            std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
    ])
    
    total_dataset = torchvision.datasets.ImageFolder(data_dir, transform=train_transforms)
    print(total_dataset, '\n')
    print(total_dataset.class_to_idx, '\n')
    
    # 划分训练集与测试集
    train_size = int(0.8 * len(total_dataset))
    test_size  = len(total_dataset) - train_size
    print('train_size', train_size, 'test_size', test_size, '\n')
    train_dataset, test_dataset = torch.utils.data.random_split(total_dataset, [train_size, test_size])
    
    return classeNames, train_dataset, test_dataset


''' 加载数据，并设置batch_size '''
def loadData(train_ds, test_ds, batch_size=32, root='', show_flag=False):
    # 从 train_ds 加载训练集
    train_dl = torch.utils.data.DataLoader(train_ds,
                                           batch_size=batch_size,
                                           shuffle=True,
                                           num_workers=1)
    # 从 test_ds 加载测试集
    test_dl  = torch.utils.data.DataLoader(test_ds,
                                           batch_size=batch_size,
                                           shuffle=True,
                                           num_workers=1)
    
    # 取一个批次查看数据格式
    # 数据的shape为：[batch_size, channel, height, weight]
    # 其中batch_size为自己设定，channel，height和weight分别是图片的通道数，高度和宽度。
    for X, y in test_dl:
        print('Shape of X [N, C, H, W]: ', X.shape)
        print('Shape of y: ', y.shape, y.dtype, '\n')
        break
    
    imgs, labels = next(iter(train_dl))
    print('Image shape: ', imgs.shape, '\n')
    # torch.Size([32, 3, 224, 224])  # 所有数据集中的图像都是224*224的RGB图
    displayData(imgs, root, show_flag)
    return train_dl, test_dl


batch_size = 4
data_dir = './data/weather_photos/'
train_ds, test_ds = localDataset(data_dir)
train_dl, test_dl = loadData(train_ds, test_ds, batch_size, data_dir, True)

Dataset ImageFolder
    Number of datapoints: 1125
    Root location: data\weather_photos
    StandardTransform
Transform: Compose(
               Resize(size=[224, 224], interpolation=bilinear)
               ToTensor()
               Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
           )

{'cloudy': 0, 'rain': 1, 'shine': 2, 'sunrise': 3}

train_size 900 test_size 225

num_classes 4
Shape of X [N, C, H, W]:  torch.Size([4, 3, 224, 224])
Shape of y:  torch.Size([4]) torch.int64

Image shape:  torch.Size([4, 3, 224, 224])

---

3.数据可视化

''' 数据可视化 '''
def displayData(imgs, root='', flag=False):
    # 指定图片大小，图像大小为20宽、5高的绘图(单位为英寸inch)
    plt.figure('Data Visualization', figsize=(20, 5)) 
    for i, imgs in enumerate(imgs[:20]):
        # 维度顺序调整 [3, 224, 224]->[224, 224, 3]
        npimg = imgs.numpy().transpose((1, 2, 0))
        # 将整个figure分成2行10列，绘制第i+1个子图。
        plt.subplot(2, 10, i+1)
        plt.imshow(npimg)  # cmap=plt.cm.binary
        plt.axis('off')
    plt.savefig(os.path.join(root, 'DatasetDisplay.png'))
    if flag:
        plt.show()
    else:
        plt.close('all')

---

三、搭建包含Backbone模块的模型

import torch
import torch.nn as nn
import torch.nn.functional as F
import torchsummary


''' 搭建包含Backbone模块的模型 '''
def autopad(k, p=None):  # kernel, padding
    # Pad to 'same'
    if p is None:
        p = k // 2 if isinstance(k, int) else [x // 2 for x in k]  # auto-pad
    return p


class Conv(nn.Module):
    # Standard convolution
    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):  # ch_in, ch_out, kernel, stride, padding, groups
        super().__init__()
        self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False)
        self.bn = nn.BatchNorm2d(c2)
        self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
    
    def forward(self, x):
        return self.act(self.bn(self.conv(x)))


class Bottleneck(nn.Module):
    # Standard bottleneck
    def __init__(self, c1, c2, shortcut=True, g=1, e=0.5):  # ch_in, ch_out, shortcut, groups, expansion
        super().__init__()
        c_ = int(c2 * e)  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c_, c2, 3, 1, g=g)
        self.add = shortcut and c1 == c2
    
    def forward(self, x):
        return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))


class C3(nn.Module):
    # CSP Bottleneck with 3 convolutions
    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):  # ch_in, ch_out, number, shortcut, groups, expansion
        super().__init__()
        c_ = int(c2 * e)  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c1, c_, 1, 1)
        self.cv3 = Conv(2 * c_, c2, 1)  # act=FReLU(c2)
        self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
    
    def forward(self, x):
        return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1))


class SPPF(nn.Module):
    # Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher
    def __init__(self, c1, c2, k=5):  # equivalent to SPP(k=(5, 9, 13))
        super().__init__()
        c_ = c1 // 2  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c_ * 4, c2, 1, 1)
        self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2)
    
    def forward(self, x):
        x = self.cv1(x)
        with warnings.catch_warnings():
            warnings.simplefilter('ignore')  # suppress torch 1.9.0 max_pool2d() warning
            y1 = self.m(x)
            y2 = self.m(y1)
            return self.cv2(torch.cat([x, y1, y2, self.m(y2)], 1))


'''这个是YOLOv5, 6.0版本的主干网络，这里进行复现（注：有部分删改，详细讲解将在后续进行展开）'''
class YOLOv5_backbone(nn.Module):
    def __init__(self):
        super(YOLOv5_backbone, self).__init__()
        
        self.Conv_1 = Conv(3, 64, 3, 2, 2) 
        self.Conv_2 = Conv(64, 128, 3, 2) 
        self.C3_3   = C3(128,128)
        self.Conv_4 = Conv(128, 256, 3, 2) 
        self.C3_5   = C3(256,256)
        self.Conv_6 = Conv(256, 512, 3, 2) 
        self.C3_7   = C3(512,512)
        self.Conv_8 = Conv(512, 1024, 3, 2) 
        self.C3_9   = C3(1024, 1024)
        self.SPPF   = SPPF(1024, 1024, 5)
        
        # 全连接网络层，用于分类
        self.classifier = nn.Sequential(
            nn.Linear(in_features=65536, out_features=100),
            nn.ReLU(),
            nn.Linear(in_features=100, out_features=4)
        )
    
    def forward(self, x):
        x = self.Conv_1(x)
        x = self.Conv_2(x)
        x = self.C3_3(x)
        x = self.Conv_4(x)
        x = self.C3_5(x)
        x = self.Conv_6(x)
        x = self.C3_7(x)
        x = self.Conv_8(x)
        x = self.C3_9(x)
        x = self.SPPF(x)
        
        x = torch.flatten(x, start_dim=1)
        x = self.classifier(x)
        
        return x


device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using {} device\n".format(device))
''' 调用并将模型转移到GPU中（我们模型运行均在GPU中进行） '''
model = YOLOv5_backbone().to(device)
''' 显示网络结构 '''
torchsummary.summary(model, (3, 224, 224))
print(model)

Using cuda device

----------------------------------------------------------------
        Layer (type)               Output Shape         Param #
================================================================
            Conv2d-1         [-1, 64, 113, 113]           1,728
       BatchNorm2d-2         [-1, 64, 113, 113]             128
              SiLU-3         [-1, 64, 113, 113]               0
              Conv-4         [-1, 64, 113, 113]               0
            Conv2d-5          [-1, 128, 57, 57]          73,728
       BatchNorm2d-6          [-1, 128, 57, 57]             256
              SiLU-7          [-1, 128, 57, 57]               0
              Conv-8          [-1, 128, 57, 57]               0
            Conv2d-9           [-1, 64, 57, 57]           8,192
      BatchNorm2d-10           [-1, 64, 57, 57]             128
             SiLU-11           [-1, 64, 57, 57]               0
             Conv-12           [-1, 64, 57, 57]               0
           Conv2d-13           [-1, 64, 57, 57]           4,096
      BatchNorm2d-14           [-1, 64, 57, 57]             128
             SiLU-15           [-1, 64, 57, 57]               0
             Conv-16           [-1, 64, 57, 57]               0
           Conv2d-17           [-1, 64, 57, 57]          36,864
      BatchNorm2d-18           [-1, 64, 57, 57]             128
             SiLU-19           [-1, 64, 57, 57]               0
             Conv-20           [-1, 64, 57, 57]               0
       Bottleneck-21           [-1, 64, 57, 57]               0
           Conv2d-22           [-1, 64, 57, 57]           8,192
      BatchNorm2d-23           [-1, 64, 57, 57]             128
             SiLU-24           [-1, 64, 57, 57]               0
             Conv-25           [-1, 64, 57, 57]               0
           Conv2d-26          [-1, 128, 57, 57]          16,384
      BatchNorm2d-27          [-1, 128, 57, 57]             256
             SiLU-28          [-1, 128, 57, 57]               0
             Conv-29          [-1, 128, 57, 57]               0
               C3-30          [-1, 128, 57, 57]               0
           Conv2d-31          [-1, 256, 29, 29]         294,912
      BatchNorm2d-32          [-1, 256, 29, 29]             512
             SiLU-33          [-1, 256, 29, 29]               0
             Conv-34          [-1, 256, 29, 29]               0
           Conv2d-35          [-1, 128, 29, 29]          32,768
      BatchNorm2d-36          [-1, 128, 29, 29]             256
             SiLU-37          [-1, 128, 29, 29]               0
             Conv-38          [-1, 128, 29, 29]               0
           Conv2d-39          [-1, 128, 29, 29]          16,384
      BatchNorm2d-40          [-1, 128, 29, 29]             256
             SiLU-41          [-1, 128, 29, 29]               0
             Conv-42          [-1, 128, 29, 29]               0
           Conv2d-43          [-1, 128, 29, 29]         147,456
      BatchNorm2d-44          [-1, 128, 29, 29]             256
             SiLU-45          [-1, 128, 29, 29]               0
             Conv-46          [-1, 128, 29, 29]               0
       Bottleneck-47          [-1, 128, 29, 29]               0
           Conv2d-48          [-1, 128, 29, 29]          32,768
      BatchNorm2d-49          [-1, 128, 29, 29]             256
             SiLU-50          [-1, 128, 29, 29]               0
             Conv-51          [-1, 128, 29, 29]               0
           Conv2d-52          [-1, 256, 29, 29]          65,536
      BatchNorm2d-53          [-1, 256, 29, 29]             512
             SiLU-54          [-1, 256, 29, 29]               0
             Conv-55          [-1, 256, 29, 29]               0
               C3-56          [-1, 256, 29, 29]               0
           Conv2d-57          [-1, 512, 15, 15]       1,179,648
      BatchNorm2d-58          [-1, 512, 15, 15]           1,024
             SiLU-59          [-1, 512, 15, 15]               0
             Conv-60          [-1, 512, 15, 15]               0
           Conv2d-61          [-1, 256, 15, 15]         131,072
      BatchNorm2d-62          [-1, 256, 15, 15]             512
             SiLU-63          [-1, 256, 15, 15]               0
             Conv-64          [-1, 256, 15, 15]               0
           Conv2d-65          [-1, 256, 15, 15]          65,536
      BatchNorm2d-66          [-1, 256, 15, 15]             512
             SiLU-67          [-1, 256, 15, 15]               0
             Conv-68          [-1, 256, 15, 15]               0
           Conv2d-69          [-1, 256, 15, 15]         589,824
      BatchNorm2d-70          [-1, 256, 15, 15]             512
             SiLU-71          [-1, 256, 15, 15]               0
             Conv-72          [-1, 256, 15, 15]               0
       Bottleneck-73          [-1, 256, 15, 15]               0
           Conv2d-74          [-1, 256, 15, 15]         131,072
      BatchNorm2d-75          [-1, 256, 15, 15]             512
             SiLU-76          [-1, 256, 15, 15]               0
             Conv-77          [-1, 256, 15, 15]               0
           Conv2d-78          [-1, 512, 15, 15]         262,144
      BatchNorm2d-79          [-1, 512, 15, 15]           1,024
             SiLU-80          [-1, 512, 15, 15]               0
             Conv-81          [-1, 512, 15, 15]               0
               C3-82          [-1, 512, 15, 15]               0
           Conv2d-83           [-1, 1024, 8, 8]       4,718,592
      BatchNorm2d-84           [-1, 1024, 8, 8]           2,048
             SiLU-85           [-1, 1024, 8, 8]               0
             Conv-86           [-1, 1024, 8, 8]               0
           Conv2d-87            [-1, 512, 8, 8]         524,288
      BatchNorm2d-88            [-1, 512, 8, 8]           1,024
             SiLU-89            [-1, 512, 8, 8]               0
             Conv-90            [-1, 512, 8, 8]               0
           Conv2d-91            [-1, 512, 8, 8]         262,144
      BatchNorm2d-92            [-1, 512, 8, 8]           1,024
             SiLU-93            [-1, 512, 8, 8]               0
             Conv-94            [-1, 512, 8, 8]               0
           Conv2d-95            [-1, 512, 8, 8]       2,359,296
      BatchNorm2d-96            [-1, 512, 8, 8]           1,024
             SiLU-97            [-1, 512, 8, 8]               0
             Conv-98            [-1, 512, 8, 8]               0
       Bottleneck-99            [-1, 512, 8, 8]               0
          Conv2d-100            [-1, 512, 8, 8]         524,288
     BatchNorm2d-101            [-1, 512, 8, 8]           1,024
            SiLU-102            [-1, 512, 8, 8]               0
            Conv-103            [-1, 512, 8, 8]               0
          Conv2d-104           [-1, 1024, 8, 8]       1,048,576
     BatchNorm2d-105           [-1, 1024, 8, 8]           2,048
            SiLU-106           [-1, 1024, 8, 8]               0
            Conv-107           [-1, 1024, 8, 8]               0
              C3-108           [-1, 1024, 8, 8]               0
          Conv2d-109            [-1, 512, 8, 8]         524,288
     BatchNorm2d-110            [-1, 512, 8, 8]           1,024
            SiLU-111            [-1, 512, 8, 8]               0
            Conv-112            [-1, 512, 8, 8]               0
       MaxPool2d-113            [-1, 512, 8, 8]               0
       MaxPool2d-114            [-1, 512, 8, 8]               0
       MaxPool2d-115            [-1, 512, 8, 8]               0
          Conv2d-116           [-1, 1024, 8, 8]       2,097,152
     BatchNorm2d-117           [-1, 1024, 8, 8]           2,048
            SiLU-118           [-1, 1024, 8, 8]               0
            Conv-119           [-1, 1024, 8, 8]               0
            SPPF-120           [-1, 1024, 8, 8]               0
          Linear-121                  [-1, 100]       6,553,700
            ReLU-122                  [-1, 100]               0
          Linear-123                    [-1, 4]             404
================================================================
Total params: 21,729,592
Trainable params: 21,729,592
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 137.59
Params size (MB): 82.89
Estimated Total Size (MB): 221.06
----------------------------------------------------------------
YOLOv5_backbone(
  (Conv_1): Conv(
    (conv): Conv2d(3, 64, kernel_size=(3, 3), stride=(2, 2), padding=(2, 2), bias=False)
    (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (Conv_2): Conv(
    (conv): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_3): C3(
    (cv1): Conv(
      (conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (Conv_4): Conv(
    (conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_5): C3(
    (cv1): Conv(
      (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (Conv_6): Conv(
    (conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_7): C3(
    (cv1): Conv(
      (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (Conv_8): Conv(
    (conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_9): C3(
    (cv1): Conv(
      (conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(1024, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (SPPF): SPPF(
    (cv1): Conv(
      (conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(2048, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): MaxPool2d(kernel_size=5, stride=1, padding=2, dilation=1, ceil_mode=False)
  )
  (classifier): Sequential(
    (0): Linear(in_features=65536, out_features=100, bias=True)
    (1): ReLU()
    (2): Linear(in_features=100, out_features=4, bias=True)
  )
)

---

四、训练模型

1.编写训练函数

optimizer.zero_grad()
loss.backward()
optimizer.step()
关于以上三个函数，我在之前的文章中有做说明，这里不再赘述

# 训练循环
def train(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)  # 训练集的大小
    num_batches = len(dataloader)   # 批次数目

    train_loss, train_acc = 0, 0  # 初始化训练损失和正确率
    
    for X, y in dataloader:  # 获取图片及其标签
        X, y = X.to(device), y.to(device)
        
        # 计算预测误差
        pred = model(X)          # 网络输出
        loss = loss_fn(pred, y)  # 计算网络输出和真实值之间的差距，targets为真实值，计算二者差值即为损失
        
        # 反向传播
        optimizer.zero_grad()  # grad属性归零
        loss.backward()        # 反向传播
        optimizer.step()       # 每一步自动更新
        
        # 记录acc与loss
        train_acc  += (pred.argmax(1) == y).type(torch.float).sum().item()
        train_loss += loss.item()
            
    train_acc  /= size
    train_loss /= num_batches

    return train_acc, train_loss

---

2.编写测试函数

测试函数和训练函数大致相同，但是由于不进行梯度下降对网络权重进行更新，所以不需要传入优化器

def test (dataloader, model, loss_fn):
    size        = len(dataloader.dataset)  # 测试集的大小
    num_batches = len(dataloader)          # 批次数目
    test_loss, test_acc = 0, 0
    
    # 当不进行训练时，停止梯度更新，节省计算内存消耗
    with torch.no_grad():
        for imgs, target in dataloader:
            imgs, target = imgs.to(device), target.to(device)
            
            # 计算loss
            target_pred = model(imgs)
            loss        = loss_fn(target_pred, target)
            
            test_loss += loss.item()
            test_acc  += (target_pred.argmax(1) == target).type(torch.float).sum().item()

    test_acc  /= size
    test_loss /= num_batches

    return test_acc, test_loss

---

3.正式训练

model.train()
model.eval()

关于以上两个个函数，我在之前的文章中有做说明，这里不再赘述

start_epoch = 0
epochs      = 50
learn_rate  = 1e-4 # 初始学习率
loss_fn     = nn.CrossEntropyLoss()  # 创建损失函数
#optimizer   = torch.optim.SGD(model.parameters(), lr=learn_rate)
optimizer   = torch.optim.Adam(model.parameters(),lr=learn_rate)
# 调用官方动态学习率接口时使用
#lambda1 = lambda epoch: 0.92 ** (epoch // 4)
#scheduler   = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda1)  # 选定调整方法

train_loss  = []
train_acc   = []
test_loss   = []
test_acc    = []
epoch_best_acc = 0

''' 加载之前保存的模型 '''
if not os.path.exists(output) or not os.path.isdir(output):
    os.makedirs(output)
if start_epoch > 0:
    resumeFile = os.path.join(output, 'epoch'+str(start_epoch)+'.pkl')
    if not os.path.exists(resumeFile) or not os.path.isfile(resumeFile):
        start_epoch = 0
    else:
        model.load_state_dict(torch.load(resumeFile))  # 加载模型参数

''' 加载之前保存的模型 '''
    if not os.path.exists(output) or not os.path.isdir(output):
        os.makedirs(output)
    if start_epoch > 0:
        resumeFile = os.path.join(output, 'epoch'+str(start_epoch)+'.pkl')
        if not os.path.exists(resumeFile) or not os.path.isfile(resumeFile):
            start_epoch = 0
        else:
            model.load_state_dict(torch.load(resumeFile))  # 加载模型参数
    
''' 开始训练模型 '''
print('\nStart training...')
best_model = None
for epoch in range(start_epoch, epochs):
    # 更新学习率（使用自定义学习率时使用）
    # adjust_learning_rate(optimizer, epoch, learn_rate)
    
    model.train()
    epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)
    # scheduler.step() # 更新学习率（调用官方动态学习率接口时使用）
    
    model.eval()
    epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
    
    train_acc.append(epoch_train_acc)
    train_loss.append(epoch_train_loss)
    test_acc.append(epoch_test_acc)
    test_loss.append(epoch_test_loss)
    
    # 获取当前的学习率
    lr = optimizer.state_dict()['param_groups'][0]['lr']
    
    template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr:{:.2E}')
    print(time.strftime('[%Y-%m-%d %H:%M:%S]'), template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss, lr))
    
    # 保存最佳模型
    if epoch_test_acc>epoch_best_acc:
        ''' 保存最优模型参数 '''
        epoch_best_acc = epoch_test_acc
        best_model = copy.deepcopy(model)
        print(('acc = {:.1f}%, saving model to best.pkl').format(epoch_best_acc*100))
        saveFile = os.path.join(output, 'best.pkl')
        torch.save(best_model.state_dict(), saveFile)
    if epoch_test_acc==1 and epoch_train_acc==1:
        saveFile = os.path.join(output, 'epoch'+str(epoch+1)+'.pkl')
        torch.save(model.state_dict(), saveFile)
print('Done\n')

''' 保存模型参数 '''
saveFile = os.path.join(output, 'epoch'+str(epochs)+'.pkl')
torch.save(model.state_dict(), saveFile)

Start training...
[2022-11-24 19:08:26] Epoch: 1, Train_acc:53.6%, Train_loss:1.168, Test_acc:78.2%, Test_loss:0.628, Lr:1.00E-04
acc = 78.2%, saving model to best.pkl
[2022-11-24 19:08:42] Epoch: 2, Train_acc:67.0%, Train_loss:0.797, Test_acc:73.8%, Test_loss:0.630, Lr:1.00E-04
[2022-11-24 19:09:00] Epoch: 3, Train_acc:76.8%, Train_loss:0.597, Test_acc:86.2%, Test_loss:0.362, Lr:1.00E-04
acc = 86.2%, saving model to best.pkl
[2022-11-24 19:09:17] Epoch: 4, Train_acc:77.9%, Train_loss:0.587, Test_acc:84.0%, Test_loss:0.428, Lr:1.00E-04
[2022-11-24 19:09:34] Epoch: 5, Train_acc:81.0%, Train_loss:0.482, Test_acc:87.6%, Test_loss:0.406, Lr:1.00E-04
acc = 87.6%, saving model to best.pkl
[2022-11-24 19:09:51] Epoch: 6, Train_acc:82.1%, Train_loss:0.488, Test_acc:85.8%, Test_loss:0.354, Lr:1.00E-04
[2022-11-24 19:10:07] Epoch: 7, Train_acc:85.7%, Train_loss:0.377, Test_acc:84.4%, Test_loss:0.382, Lr:1.00E-04
[2022-11-24 19:10:23] Epoch: 8, Train_acc:88.3%, Train_loss:0.316, Test_acc:90.7%, Test_loss:0.280, Lr:1.00E-04
acc = 90.7%, saving model to best.pkl
[2022-11-24 19:10:39] Epoch: 9, Train_acc:89.9%, Train_loss:0.276, Test_acc:94.2%, Test_loss:0.192, Lr:1.00E-04
acc = 94.2%, saving model to best.pkl
[2022-11-24 19:10:56] Epoch:10, Train_acc:88.3%, Train_loss:0.315, Test_acc:93.3%, Test_loss:0.227, Lr:1.00E-04
[2022-11-24 19:11:13] Epoch:11, Train_acc:88.2%, Train_loss:0.304, Test_acc:93.3%, Test_loss:0.162, Lr:1.00E-04
[2022-11-24 19:11:29] Epoch:12, Train_acc:90.6%, Train_loss:0.254, Test_acc:94.7%, Test_loss:0.165, Lr:1.00E-04
acc = 94.7%, saving model to best.pkl
[2022-11-24 19:11:45] Epoch:13, Train_acc:93.3%, Train_loss:0.193, Test_acc:96.4%, Test_loss:0.117, Lr:1.00E-04
acc = 96.4%, saving model to best.pkl
[2022-11-24 19:12:01] Epoch:14, Train_acc:94.4%, Train_loss:0.172, Test_acc:92.0%, Test_loss:0.231, Lr:1.00E-04
[2022-11-24 19:12:18] Epoch:15, Train_acc:93.6%, Train_loss:0.168, Test_acc:90.7%, Test_loss:0.243, Lr:1.00E-04
[2022-11-24 19:12:34] Epoch:16, Train_acc:95.8%, Train_loss:0.126, Test_acc:92.4%, Test_loss:0.236, Lr:1.00E-04
[2022-11-24 19:12:50] Epoch:17, Train_acc:95.1%, Train_loss:0.129, Test_acc:92.0%, Test_loss:0.284, Lr:1.00E-04
[2022-11-24 19:13:06] Epoch:18, Train_acc:97.1%, Train_loss:0.079, Test_acc:95.6%, Test_loss:0.123, Lr:1.00E-04
[2022-11-24 19:13:22] Epoch:19, Train_acc:95.3%, Train_loss:0.135, Test_acc:92.4%, Test_loss:0.196, Lr:1.00E-04
[2022-11-24 19:13:37] Epoch:20, Train_acc:94.9%, Train_loss:0.149, Test_acc:90.2%, Test_loss:0.325, Lr:1.00E-04
[2022-11-24 19:13:53] Epoch:21, Train_acc:95.3%, Train_loss:0.137, Test_acc:94.7%, Test_loss:0.212, Lr:1.00E-04
[2022-11-24 19:14:09] Epoch:22, Train_acc:96.7%, Train_loss:0.115, Test_acc:94.7%, Test_loss:0.161, Lr:1.00E-04
[2022-11-24 19:14:26] Epoch:23, Train_acc:98.6%, Train_loss:0.051, Test_acc:95.1%, Test_loss:0.173, Lr:1.00E-04
[2022-11-24 19:14:43] Epoch:24, Train_acc:98.9%, Train_loss:0.040, Test_acc:93.8%, Test_loss:0.252, Lr:1.00E-04
[2022-11-24 19:15:06] Epoch:25, Train_acc:98.8%, Train_loss:0.049, Test_acc:95.1%, Test_loss:0.172, Lr:1.00E-04
[2022-11-24 19:15:53] Epoch:26, Train_acc:97.1%, Train_loss:0.083, Test_acc:90.2%, Test_loss:0.320, Lr:1.00E-04
[2022-11-24 19:16:38] Epoch:27, Train_acc:94.9%, Train_loss:0.133, Test_acc:90.2%, Test_loss:0.327, Lr:1.00E-04
[2022-11-24 19:17:23] Epoch:28, Train_acc:97.9%, Train_loss:0.072, Test_acc:94.2%, Test_loss:0.253, Lr:1.00E-04
[2022-11-24 19:18:08] Epoch:29, Train_acc:98.9%, Train_loss:0.032, Test_acc:88.0%, Test_loss:0.418, Lr:1.00E-04
[2022-11-24 19:18:50] Epoch:30, Train_acc:96.3%, Train_loss:0.109, Test_acc:88.0%, Test_loss:0.514, Lr:1.00E-04
[2022-11-24 19:19:33] Epoch:31, Train_acc:96.7%, Train_loss:0.098, Test_acc:90.7%, Test_loss:0.308, Lr:1.00E-04
[2022-11-24 19:20:15] Epoch:32, Train_acc:98.8%, Train_loss:0.032, Test_acc:93.3%, Test_loss:0.239, Lr:1.00E-04
[2022-11-24 19:20:57] Epoch:33, Train_acc:98.2%, Train_loss:0.047, Test_acc:94.7%, Test_loss:0.192, Lr:1.00E-04
[2022-11-24 19:21:39] Epoch:34, Train_acc:96.9%, Train_loss:0.066, Test_acc:92.9%, Test_loss:0.266, Lr:1.00E-04
[2022-11-24 19:22:21] Epoch:35, Train_acc:98.1%, Train_loss:0.045, Test_acc:93.8%, Test_loss:0.285, Lr:1.00E-04
[2022-11-24 19:23:02] Epoch:36, Train_acc:98.8%, Train_loss:0.044, Test_acc:89.8%, Test_loss:0.512, Lr:1.00E-04
[2022-11-24 19:23:44] Epoch:37, Train_acc:99.2%, Train_loss:0.028, Test_acc:92.0%, Test_loss:0.424, Lr:1.00E-04
[2022-11-24 19:24:26] Epoch:38, Train_acc:97.9%, Train_loss:0.060, Test_acc:92.4%, Test_loss:0.283, Lr:1.00E-04
[2022-11-24 19:25:09] Epoch:39, Train_acc:98.4%, Train_loss:0.039, Test_acc:91.1%, Test_loss:0.272, Lr:1.00E-04
[2022-11-24 19:25:51] Epoch:40, Train_acc:98.0%, Train_loss:0.058, Test_acc:89.3%, Test_loss:0.360, Lr:1.00E-04
[2022-11-24 19:26:33] Epoch:41, Train_acc:97.3%, Train_loss:0.070, Test_acc:92.0%, Test_loss:0.394, Lr:1.00E-04
[2022-11-24 19:27:15] Epoch:42, Train_acc:98.9%, Train_loss:0.031, Test_acc:93.3%, Test_loss:0.345, Lr:1.00E-04
[2022-11-24 19:27:58] Epoch:43, Train_acc:98.9%, Train_loss:0.022, Test_acc:94.7%, Test_loss:0.332, Lr:1.00E-04
[2022-11-24 19:28:40] Epoch:44, Train_acc:98.9%, Train_loss:0.037, Test_acc:93.3%, Test_loss:0.251, Lr:1.00E-04
[2022-11-24 19:29:22] Epoch:45, Train_acc:97.2%, Train_loss:0.078, Test_acc:92.9%, Test_loss:0.324, Lr:1.00E-04
[2022-11-24 19:30:04] Epoch:46, Train_acc:98.8%, Train_loss:0.033, Test_acc:92.9%, Test_loss:0.389, Lr:1.00E-04
[2022-11-24 19:30:47] Epoch:47, Train_acc:99.4%, Train_loss:0.025, Test_acc:91.1%, Test_loss:0.416, Lr:1.00E-04
[2022-11-24 19:31:29] Epoch:48, Train_acc:98.8%, Train_loss:0.033, Test_acc:93.8%, Test_loss:0.358, Lr:1.00E-04
[2022-11-24 19:32:11] Epoch:49, Train_acc:99.4%, Train_loss:0.013, Test_acc:93.8%, Test_loss:0.336, Lr:1.00E-04
[2022-11-24 19:32:53] Epoch:50, Train_acc:99.9%, Train_loss:0.011, Test_acc:93.8%, Test_loss:0.321, Lr:1.00E-04
Done

最终结果，在第23轮时(Epoch:13的结果)的训练集准确率达到93.3%，测试集准确率达到96.4%。

---

五、结果可视化&模型评估

1.训练结果可视化

import matplotlib.pyplot as plt
import warnings


''' 结果可视化 '''
def displayResult(train_acc, test_acc, train_loss, test_loss, start_epoch, epochs, output=''):
    # 隐藏警告
    warnings.filterwarnings("ignore")                # 忽略警告信息
    plt.rcParams['font.sans-serif']    = ['SimHei']  # 用来正常显示中文标签
    plt.rcParams['axes.unicode_minus'] = False       # 用来正常显示负号
    plt.rcParams['figure.dpi']         = 100         # 分辨率
    
    epochs_range = range(start_epoch, epochs)
    
    plt.figure('Result Visualization', figsize=(12, 3))
    plt.subplot(1, 2, 1)
    
    plt.plot(epochs_range, train_acc, label='Training Accuracy')
    plt.plot(epochs_range, test_acc, label='Test Accuracy')
    plt.legend(loc='lower right')
    plt.title('Training and Validation Accuracy')
    
    plt.subplot(1, 2, 2)
    plt.plot(epochs_range, train_loss, label='Training Loss')
    plt.plot(epochs_range, test_loss, label='Test Loss')
    plt.legend(loc='upper right')
    plt.title('Training and Validation Loss')
    plt.savefig(os.path.join(output, 'AccuracyLoss.png'))
    plt.show()


''' 绘制准确率&损失率曲线图 '''
displayResult(train_acc, test_acc, train_loss, test_loss, start_epoch, epochs, output)

---

2.模型评估

best_model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)
print("EVAL {:.5f}, {:.5f}".format(epoch_test_acc, epoch_test_loss))

EVAL 0.96444, 0.11800

---